I’ve covered data cubes in a previous three-part series on business intelligence, but I wanted to spend some time explaining how they play into data management and your ability to find the answers you need.
As I’ve explained, data cubes consolidate data from multiple sources so that complicated answers can be computed in advance. The impact of this advance calculation, as we’ve seen first-hand for one client as we’ve worked to improve their business intelligence capabilities, is that processes that used to require whole days spinning data in Excel now occurs in a matter of minutes.
While this new process and the data cube itself requires little to no effort, to the point that changes can be made using drag and drop functionality, the actual architecture behind this simplicity is anything but. More importantly, getting this structure set up correctly from the start, by a software consultant who understands your business needs, is vital to the success of your data management project.
To start with, it takes an expert that understands data management to create an effective system for feeding the data cube. The basics of defining data sources, transforming them into the model you want, and loading them into the cube must be done right for it to be able to answer your questions correctly.
Two areas that can make a big difference in the efficiency of your data cube are defining timing and the roles that the data within the cube will address. In terms of timing, it is possible for data cubes to process on a minute by minute or even second by second basis. But most organizations simply don’t have the need for that level of granularity. The more your data cube has to process, the slower it will work. So take a good look at the answers you’re looking to get back for before you decide whether hourly or daily processing might meet the need.
The other area that can make a difference is when you create data cubes by role. For example, a sales manager might need to see more current information than an accountant in order to be effective at their job. This will help define a processing schedule for the data cube, based on urgency, that guarantees maximum effectiveness.
For more on how data cubes translate into valuable business intelligence that you can use for better decision making, read the complete series on the components of a business intelligence solution.